Open Access Open Access  Restricted Access Subscription Access

Copyright © 2021 ISEIS. All rights reserved

Spatial Distribution Patterns of Eurasian Otter (Lutra Lutra) in Association with Environmental Factors Unravelled by Machine Learning and Diffusion Kernel Method

S. Hong1,2, T.-S. Chon3,4 *, and G. J. Joo5 *

  1. Department of Animal Science and Biotechnology, Kyungpook National University, Sangju 37224, Republic of Korea
  2. Department of Horse, Companion, and Wild Animal Science, Kyungpook National University, Sangju 37224, Republic of Korea
  3. Department of Forest Environment Protection, College of Forest and Environmental Sciences, Kangwon National University, Chuncheon 25931, Republic of Korea
  4. Ecology and Future Research Association, Busan 46228, Republic of Korea
  5. Department of Biological Sciences, Pusan National University, Busan 46241, Republic of Korea

*Corresponding author. Tel.: +82-51-512-2262. E-mail address: (T.-S. Chon).
*Corresponding author. Tel.: +82-51-510-2258; fax: +82-51-581-2962. E-mail address: (G. J. Joo).


In South Korea, the endangered Eurasian otter (Lutra lutra) populations have been recovered throughout the country. To examine the status of otter populations, we monitored spraint densities at 250 ~ 355 sites annually from 2014 to 2017 in the Nakdong River basin. The diffusion kernel method was applied to both binary and continuous spraint data. Two geographical popula - tions were identified: northern and southern populations. The northern population continuously increased over a broad area from north to south during the study period. In contrast, the southern population narrowly dispersed, limited by its location in an industrial area. The spatial self-organizing map (Geo-SOM) revealed associations between spraint densities and environmental factors by correlating the geographic locations of the sampling sites. Both populations were negatively affected by anthropogenic factors, including proximi - ty to factories and human population density. However, cumulative association of all environmental factors, including landscape, food sources, and anthropogenic factors, were noted in 2016 after which otter populations fully recovered. Population development stabilized while exhibiting an overall high association with environmental factors. The results of the diffusion kernel method and data variation according to the Geo-SOM consistently presented substantial change in population dispersal (i.e. the merge of two subpopulations, and complete associations between spraint and environmental factors). The combination of the diffusion kernel method and Geo-SOM was effective in portraying temporal changes in population states in association with environmental factors based on spra int data in the last phase of full recovery.

Keywords: Dispersal; recovery; spraints; habitat preference; conservation; Republic of Korea

Full Text:


Supplementary Files:


  • There are currently no refbacks.